Notes on Engineering Health, August 2021
Evolution and Public Health
What does it take for a virus to create new variants and to evolve into a more deadly foe?
If we were to model the probability of a virus gaining a function — e.g., escaping a vaccine, increasing its virulence, or being more harmful to children — in the simplest possible way, we would describe it as a function of the virus’s mutation rate (M), its replication rate (R), and the evolutionary pressure applied to it (P) — the evolutionary, or selection, pressure being any change in the environment that advantages particular mutations in the virus that are passed on to the next generation. This simple probability function would look like this:
P(Gain of function)=f(M, R, P)
R in these models is composed of both the replication rate of the virus in its host (Rh) and its reproduction rate from host-to-host, the now infamous R0. Breaking these two variables out, the probability function then gets only mildly more complicated:
P(Gain of function)=f(M, Rh, R0, P)
Experimentally, the rates of mutation (M) and replication within the host (Rh) are parameters that are challenging to change as they are inherent properties of the organism. M and Rh depend respectively on how accurately and quickly polymerases replicate DNA (or RNA). What is clear is that low mutation and replication rates make the probability of a gain of function lower.
P is, however, a parameter scientists can more easily control experimentally. For example, in a 2016 Science publication, a team of scientists created a striking visualization of how quickly E. coli can adapt to increasing concentrations of antibiotics. In this instance, the evolutionary pressure in the system is the antibiotic concentration and its increase by an order of magnitude at regular intervals. Depending on the desired outcome, the pressure has to be tweaked with precision: too high of a jump in concentration, the bacteria won’t be able to adapt and will be eradicated; too low of a jump, adaptation might not be necessary and a gain of function won’t occur.
If we turn our attention to the COVID-19 epidemic, there is no doubt that some collective and individual actions can reduce the probability of producing a more deadly variant: reducing R0 by masking, and increasing P enough to eradicate the virus by getting vaccinated are the most obvious methods. The US (and the entire world really) are currently in a dangerous phase where the vaccination rate is high enough to promote the selection of mutations that will evade the vaccine’s efficacy but still too low to eradicate the virus entirely. It is important to note that the effect of vaccination on the ability of the virus to spread (and thus mutate) is hard to measure but is clearly not linear. According to data from ourworldindata.org and recent models, the highest risk for vaccine escape occurs at intermediate levels of vaccination. Not getting a vaccine when the opportunity presents itself and entrenching uneven access to vaccines across the globe boost the emergence of new variants, and the likelihood that one of them will eventually evade the existing vaccines is all but certain.
The giant evolutionary experiment we are collectively, and largely unwillingly, part of is following well-studied rules and everyone should understand them. If you want to read further, here are good recent articles from STAT and the The Atlantic that go over these rules, and a PLoS article that goes into greater technical detail.
First Five is our list of essential media for the month which spans a range of content including scientific papers, books, podcasts, and videos. For our full list of interesting media in health, science, and technology, updated regularly, follow us on Twitter or Instagram.
1/ Incompleteness of Biology
Interesting speculation on Incompleteness in Biology and its Implications for Bioengineering. How do the ideas underlying Kurt Godel’s incompleteness theorem apply in the biology context?
2/ Fat and Metabolism
Lots of interesting science in the last month related to fat and metabolism:
– It is well known that fat cells can influence our sensitivity to insulin. A recent Cell Metabolism paper identifies three different subtypes of mature fat cells in white adipose tissue and finds that only one (AdipoPLIN) responds to insulin. The findings may be relevant for future treatments of metabolic diseases such as Type 2 diabetes.
– Beige fat seems to lower inflammation associated with the more common white fat, and thus provides some protection from dementia according to a study in Nature Communications.
– A study published in the FASEB Journal found that resistance training regulates fat cell metabolism at a molecular level.
– Daily energy expenditures fluctuate over the course of a normal human life according to a new Science paper, with clear implications for how we should manage nutrition at various ages. The New York Times wrote about the findings here.
3/ What should you eat for dinner?
Data confirms that herbs and spices are good, but there is little evidence that spicy food in hot countries is an adaptation to reducing infection risk. All things fermented are good, period. Eating flavonoid-rich foods seems to be good for your blood pressure. Diets rich in vitamin K seem to significantly protect against atherosclerosis-related cardiovascular disease. What time of day you eat your spicy fermented snacks can have a big impact as well — protein metabolism varies depending on the body’s internal biological clock. A paper on “chrononutrition” found that intake of proteins at breakfast increased muscle size and function in mice and humans.
4/ What shouldn’t you eat, probably ever?
Ultraprocessed foods are bad, and distressingly make up 2/3 of the calories in the diets of US children and teens. And the choices you make in your diet can add or subtract minutes, hours, and years of life.
5/ When I’m Sixty Four
Science continues to drive forward our understanding of aging while highlighting the challenges inherent in our lack of understanding:
– Microbes in the gut can reverse aspects of aging-related deterioration in the brain and cognitive function. Nature Aging >
– Researchers purport to have created an inflammatory clock of aging (iAge) which measures inflammatory load and predicts multi-morbidity, frailty, immune health, cardiovascular aging and is also associated with exceptional longevity in centenarians. Nature Aging >
– Global dementia cases are forecasted to triple by 2050. Alzheimer’s & Dementia >
Participated in Series A funding led by RiverVest Venture Partners and Mass General Brigham Ventures, with additional participation from Casdin Capital, KdT Ventures, Bold Capital Partners, and Alexandria Venture Investments.
SeQure Dx is applying state-of-the-art genomics technologies for gene editor target profiling to advance therapeutic discovery, clinical development, and diagnostic solutions. The company’s core technologies enable identification of all potential off-target sites prior to initiating therapy, allowing comprehensive confirmation of actual edits. This information provides gene editing therapeutics developers, physicians, and patients the insights needed to ensure safe and effective treatments.
UnityPoint Health launches social services tool powered by Aunt Bertha
The Gazette >
Maternal health platform Cayaba Care launches with $3.2 million in seed funding
Dataset on core body temperature reaches record size, leveraged to help detect heat risk
Modern Materials Handling >
Second Genome Announces Milestone Accomplished in Collaboration with Bayer, Demonstrating Broad Application of Company’s Machine Learning Platform to Identify Novel Proteins for the Development of Next-Generation Insect-Control Solutions
Second Genome Presents Preclinical Data Demonstrating the Effects of Microbial-Derived Metabolites in Autism Spectrum Disorder
Portfolio Job Opportunities
Digitalis Portfolio Companies Are Hiring
See Open Positions >
Public-Interest Technologies for Better Health
Digitalis Commons is a non-profit that partners with groups and individuals striving to address complex health problems by building public-interest technology solutions that are frontier-advancing, open-access, and scalable.
This piece On the Opportunities and Risks of Foundation Models is very long, but the sociotechnical issues it raises related to the continued rapid evolution of AI technologies are fundamental and in need of more work both theoretical and applied.